Analyzing correlation of urban functionality and spatial configuration

Abstract

As the urban development continues, different parts of a city can grow or decay by varying degrees due to natural deterioration or changes in governmental policies. City is the center for business, culture, and social activities; is the place that reflects geography, commerce, society, and culture context; is a complex whole that is woven by “people,” “activity,” and “space” “People” in the “space” can process “activities,” therefore, activities and space can influence each other. In this research, we aim to obtain insights on urban morphology by analyzing the changes in urban spaces’ “activity” patterns. We aim to explore the relationship between urban spatial configuration and functionality. Space syntax methodology is applied to investigate the urban spatial structure concurrently with the analytic hierarchy process method for evaluating experts’ survey responses to attain urban functionality index. Subsequently, the analytical data are compared and categorized to reveal the city districts’ spatial characteristics and their correlation. This research has proven that the analysis findings are consistent with the actual urban conditions, and thus affirms our analytical framework of having a creditable comprehension of the advantage and disadvantage of urban spatial integrations and its functionality. Therefore, our research methodology can be applied as an efficient preliminary evaluation tool for conceiving the merits of city districts. Based on the findings of our research, authorities will be able to discuss the urban development agendas in greater detail with higher efficiency.

Keywords

Space syntax urban function urban development

Introduction

Urbanicity is a social product by the interaction of human and space; humans perform activities in space; changes in urban spaces will influence activities that happen. In this research, we aim to obtain insights on urban morphology by analyzing the changes in “activity” patterns in the city. It is based on the strong interrelationship of the two that we regard activity density and intensity as good proxies for “urban functionality.” Therefore, urban morphology can be perceived by observing the changes in activities, and spatial patterns can be recognized through the human congregation.

Hillier and Hanson (1984) have proposed space syntax as a methodology of abstracting spatial configuration into simple components of nodes and connected network of lines. Over the past decades, space syntax techniques have been used commonly by city planners and extensively studied by researchers in the literature of architecture, urban design, planning, transportation, and interior design. Among the most prominent works, Peponis et al. (1997) analyzed Atlanta city’s underlying systematic spatial relationship between the pedestrian and vehicular movement and discovered when the city’s spatial configuration is more directly connected; it is more likely to attract significant movement. Koohsari et al. (2013) reveal the positive correlation between high connectivity arterial roads and the frequency of vehicular movements which results in diminishing the willingness of pedestrian crossing. Wang et al. (2008) analyzed Beijing Hutong’s spatial pattern and uncovered that the streets are an essential support system for diverse social interaction and activities. Space syntax is also being used for design as a tool; Hoeven and Akkelies(2014) apply it to make improvements in the existing subway station. Literature was also found in the context of Taiwan; Lee and Lin (2007) conducted studies to analyze Taiwan’s aboriginal village’s spatial configuration. The research confirmed that even though the methodology is originated from the western urban condition, it is also applicable in the context of Asian cities like Taiwan. Tsou and Huang (2009) have applied space syntax as one of the methodologies to understand the correlation of street network and urban commercial space distribution; the conclusion reveals legitimate resolution to reflect the actual condition. However, for further studies on the topic, the actual distance is recommended for consideration to attain higher accuracy for modeling. Therefore, space syntax has proven to be an appropriate and highly reliable tool for performing analysis on the city’s spatial configuration for Taiwan as well as abroad.

This study analyzes the correlation between urban spatial configuration and urban functionality. For comparative purposes, this research adopted the “common urban venues” concept discussed in the related literature of urban studies (Shih, 2017; Tseng, 2011) and designating the venues to be the “theaters.” Going to theatres is considered as the most accepted leisure activity at all times and could draw a large crowd to congregate. Theaters are often located on the major commercial arterial roads, or else the “theater” itself will quickly become the activator for the surrounding commercial developments. Before the rise of modern theaters, Temple’s front-court was the place to host outdoor performances, e.g. traditional Taiwanese opera or glove puppetry. Temple’s front-court is also viewed as the community center as well as marketplaces for many. This urban phenomenon can still be found today at many shriving night markets that have originated from these front-courts. Therefore, we can assume that the historical temple front-courts have the same spatial properties of today’s theater. Based on the active social characteristic and commonality through time, they are the most suitable “common urban venues” for understanding the pattern of the human congregation and commercial activities of our research. This research demonstrates a comparative analytical method through investigating the changes of urban activity patterns and spatial configuration development over time to better understand the capabilities, as well as the strength and weakness, of different urban spaces in the city.

Research methodology

Space syntax

Space syntax is a methodology first introduced by Hillier and Hanson in 1984; it is both a spatial configuration theory that expresses social meaning and an analytical tool that quantifies and visualizes the relationship between human societies and their inhabited spaces through abstract mapping techniques.Space syntax is only used to analyze the spatial pattern and environmental behaviour in the natural state and is not affected by external influences (such as policies, land prices, and comfort) (Hillier, 1996). In this research, we will be performing our analysis on Global Integration (Rn) on these spatial properties. Integration measures the degree to which a place is integrated or segregated with respect to the whole area Rn (equation (1)). These values are linked to the concept of accessibilities where the higher the value is, the higher the accessibility it represents.

Global integration (Rn)

R n = \frac{1}{{RRA}_{j}}

(1)

{RRA}_{j} = \frac{{R A}_{j}}{D_{k}}

{R A}_{j} = \frac{2 ({M D}_{j} - 1)}{n - 2}

{M D}_{i} = \frac{\sum_{j = i}^{n} d_{i j}}{n - 1}

where

d_{i j}

is the shortest path between line i and line k; n is the number of the nodes in the space network;

D_{k}

are the RA values for the diamond-shaped pattern, see table of D-Value in Hiller and Hanson(1984: 112).

Urban function index

Sawada et al. (1975) have created the index of Central Business District-ratio (CBD-ratio) for Asia’s commercial districts. The theoretical perspective is that high-end commercial service equals higher functionality index because it serves larger consumer groups from broader geographical locations where everyday commodity services score lower functionality index due to its limited clientele groups from the locals. The weighted index values are assigned for different commercial facilities after individual evaluations. Many researchers have cited this index for their research in the context of Taiwan (Huang, 1995; Liang, 1999). In our research, Urban Function is defined as land uses for supporting various social activities, e.g. residential districts, commercial districts, and open spaces, etc., and the urban function index (UFI) (equation (2)) is the quantitative evaluation of urban functionality which is determined by referencing the CBD-ratio quantification procedure. The original CBD-ratio index is based on Activity Based Classification; Sawada et al. (1975) have divided the commercial facilities into 10 groups and converted the rating from 1 to 10 in sequence. However, for researches in Taiwan context, many have referred to Huang (1995) instead who has divided the commercial facilities into five levels and rated them from 1 to 5 sequentially. Evaluation of traditional urban functions is often measured by population or number of servicing units, and it is not possible to evaluate open spaces where there is flux of population from time to time based on social events. Also, urban functions can be referred with different sizes and units (e.g. number of employment-population, resident population, or number of event participants), which would make them unfeasible to compare directly. Therefore, we are proposing to improve the original formula (2) for the four major land use categories (e.g. residential district, commercial district, public facility, open space) of Taipei city, and apply analytic hierarchy process (AHP) on invited experts’ opinions to formulate urban function’s weighted ratio.

The higher the UFI represents a higher intensity of the functionality. To be able to evaluate mix-use districts’ functionality, the analysis will be based on the Land Use Regulation’s floor area ratio (FAR) and the planned function usage instead of actual utilization condition.

UFI = \sum_{i = n}^{n} (A_{i} \times W_{i})

(2)

where Ai is i Max land use FAR ratio (%, for this research the calculation is based on Urban Land Use regulation to measure the maximum FAR ratio in percentage for all land use categories within the study block area) and Wi is i land use functionality scale ratio (for this research, scale ratio for each land use category is based on AHP expert survey outcome evaluation).

Purpose of research

The objective of this research is to gain an understanding of whether current urban planning is appropriate to support people’s need by comparing the correlation between urban spatial configuration and urban functionality. Space syntax methodology focuses on calculating positions of each different place in relation to the whole spatial network. For instance, higher Global Integration (Rn) value implies a space that has easier accessibility, better connectivity to adjacent areas, and is attracting larger crowds to congregate. However, space syntax analysis does not consider actual activities but predicts the characteristics of an urban space based on the computed findings. Although we often observe spaces in the city with a high pedestrian flow which have a higher intensity of activities. For activities to take place can be influenced by many other factors, such as governmental policies, context environment, and land-use regulations. Therefore, in order to investigate activity patterns, we are also relying on Land-use planning and the AHP to evaluate urban functionality through generating UFI.

Analysis of spatial configuration

Traditional settlements in Taiwan can generally be divided into three periods, prior to 1895 during Qing dynasty, between 1895 and 1945, and 1945 to 1990 (Lin, 2015). During the earlier settlement period (prior to 1895), the developments were growing organically and mostly originated from the community spiritual centers – Temples where each cluster of settlement slowly expands till merging with the other settlements to form the greater city. The most distinct pattern can be located at the “three market-streets” (Monga, Dadaocheng, and the Inner City) where each settlement cluster grows individually. The second period (1895 to 1945) was under the Japanese colonization, which has transformed Taiwan’s city development from organic growth to strategically planned development. In 1937, the Japanese government implemented “Taiwan Urban Planning Regulation,” which became the foundation for the subsequent development of urban planning legislation framework. By the end of the second period, the pattern of old “three market-streets” was no longer distinct. The third period (after 1945) began after the Second World War when Taiwan was returned to the governance of the Kuomintang (K.M.T.). Taipei city began to develop vastly, especially in the year of 1949, when the K.M.T. government retreated to Taiwan, with approximately 100 million immigrants from mainland China (Yin and Chen, 2016). In order to meet the vast demand of immigrants, many green open spaces were occupied by informal settlements. The lack of proper infrastructure to support the overcrowded population has resulted in a detrimental quality of life not only for the immigrants but also for the adjacent neighborhoods as the informal settlements have encroached on the neighbors’ open spaces. In the 1990s, these urban issues have raised great attention both from the government as well as the citizen, where improvement plans were undergoing to demolished the illegal settlements for redevelop greenway. For instance, the newer urban planning area (Xinyi District) is planned to accommodate a much higher FAR than the historical city districts.

In this study, Taipei city’s spatial configurations dated in 1895, 1907, 1945 and 2018 are going to be analyzed by Segmented Axial method (from Space Syntax Methodology) for global integration (Rn) (Figure 1). The “common urban venue”—Theater—will be the referencing index of the actual city developments for the different era. Going to theaters is a popular recreational activity for all times which attract large crowds and often conveniently located on the well-connected arterial roadways. In 1895, modern theaters in Taiwan had not been introduced, and Temple’s front-courts were the places that people congregate for social events and hosting performances. Historically, it is the space that serves similar functions as a theater does today. It is because of their capability of attracting crowds and being commonly identifiable throughout different points in history that we are designating the theaters (Temple’s front-courts) as the common urban venue for observation of crowd assembly and commercial activities.

Figure 1.

Global integration analysis (Rn) and theater distribution overlay.

In 1895, Taipei city’s street network was organically laid out except the inner city area (Taipei Old Town). From global integration analysis, three higher Rn value network clusters are at Monga, Dadaocheng, and the Inner City, which were known as “three market-streets” in historical studies, where these social clusters are concurrently located with the historical Temples in the area (Tseng, 2011). The significant historical Temples (Longshan Temple of Monga, Bao-an Temple of Dadaocheng, Greater Mazu Temple of Taipei city) are situated on the most integrated streets where newer Temples are found at the city periphery. Studies (Lin, 2015) had shown that historically Temples were often established before the streets were laid out. Monga is the oldest (established in 1723) cluster, followed by Dadaocheng (established in 1853), the new port village, and the last is the inner Taipei city (established in 1884), the political center for administrative functions. The top 20% of Rn analysis value is within Dadaocheng area, especially at the intersection of the three market-street clusters where the wharf-rail interchange station was located. And the lowest 20% of the Rn readings are at the north of Dadaocheng district where the most isolated and relatively newer settlements are. Therefore, our Rn value findings for 1895 are accurately corresponding with the actual utilization of the urban functions.

In 1907, during Japanese colonization, the Japanese have implemented a city rectification plan to transform Taipei city’s street network into a grid pattern. The new urban planning had also dismantled Taipei Old Town city walls but retained the footprint for building four new major arterial roads (Zhongshan South Road, Aiguo West Road, Zhonghua Road, Zhongxiao West Road). These transformations have obscured the three-market street’s distinctive boundary from 1895’s city configuration. The top 20% of the Rn value is now along the major arterial roads around the old town edge and other extended street networks from these main streets (today’s Yanping N. Road, Aiguo E. Road, Guilin Road, Zhongxiao E. Road). The lowest 20% of the Rn value is at the periphery of the city core; however, in 1907, the Rn value was not entirely corresponding to the actual urban activity pattern due to political influences. During Japanese colonization, the focused development area was only at Taipei inner city where most of the Japanese demographic lives with extensive street network system and numerous theaters were located (five found in the inner city, one in Dadaocheng, and two in Monga). The actual active settlements were located outside of the inner-city where most of the Taiwanese people live (Lin, 2015).

In 1945, Taipei city’s spatial configuration remained unchanged but had extended further outward to accommodate the urban growth. The major extensions are toward the north (North of Dadaocheng (6)), east (Housing development (7) and Official residences (8)), and southeast (Gongguan (9)); new urban functions were planned in addition to the existing programs. From the Rn finding, the most integrated streets (Rn Top10%) are today’s major arterial roads. Apart from these major roadways, the rest of the city has a relatively even Rn distribution, which implies that the centralized city development plan is diminishing. Additionally, research has also indicated that theaters were more uniformly located in all major social settlements during the latter half of the century.

In 1983, Taipei city’s spatial configuration had not deviated from the implemented city rectification plan; however, except the naturally restricted areas which were unsuitable for developments, the rest of the flatlands had saturated their development capacity. Urban growth was a result of the housing demand from the large immigrate population retreated from China with Kuomintang (KMT) after the Chinese Civil War in 1949. The immigrant population has increased 1,000,000 counts between 1949 and 1950 (Yin and Chen, 2016). Most of these immigrants congregated in the urban areas which made the population demand greatly exceed the anticipated urban capacity. Therefore, the designated land use for parks or other non-building functions has changed into buildable areas. Among our study boundary of Taipei city, there were two examples of informal settlements resulting from this particular urban condition, e.g. the Zhonghua road (1) informal settlement (designated land used for road) and No. 7 Park (2) informal settlement (designated land used for park, today’s Daan Forest Park). With the insufficient support of infrastructures and problem of overpopulation, the residents were suffering from a lack of open spaces. All of the above conditions have contributed to an overall decrease of urban living condition. In 1976, Taipei city proposed to implement the “sub-city plan” to relocate the Taipei city Government facilities to the eastern region; this has increased the speed of developments of the eastern areas. Although the eastern region had a complete street network in place at that time, most of the land parcels were still underdeveloped. The northern part of Taipei city boundary had expanded toward north-east areas where Songshan (3) is; the south-east edge had extended to Gongguan (4); the southwest part had stopped growing during Japanese colonial period but later extended to Xindian riverbank (5) where today’s south of Wanhua (known initially as Monga) (6) is sitting. From the Rn mapping of 1895, 1907, 1945, high Rn (Rn top 20%) are located along the major arterial roads and especially populated on the east–west major roadways. The Rn analysis results for 1983 and 2018 show the highest Rn (Rn top10%) areas are mainly in line with all the major orthogonal arterial roadways. The findings also indicate that highly integrated (high Rn value) streets are shifting eastward in comparison to 1945’s Rn analysis result and begin to develop multiple highly integrated centers (Taipei Main Station (7), Zhongxiao E. Road Commercial District (8), and Xinyi Plan District (9)).

Today (2018), Taipei city’s spatial configuration has remained similar to the 1983’s structure where the flatlands have saturated their development capacity. Therefore, the city’s spatial pattern and boundary have not differed much since. One significant difference is that the public open spaces have been reclaimed from the previous illegal uses – the most representative example is the Daan Forest Park (2) where the No. 7 park informal settlement was. In the 1990s, the city government and people have begun to collaborate on making improvement plans to solve the deteriorating urban conditions since 1983. In 1960, Zhonghua road (1) informal settlement had been demolished and transformed into Guanghua Digital Plaza and later again been removed to giveaway for creating the urban green corridor. No. 7 park informal settlement was also demolished as well in 1992 to return the land for creating Dann Forest park for public uses. Furthermore, there was increasing awareness to incorporate public spaces in the city planning policy. Tzeng et al. (1995) have proposed research on creating design guidelines of open green parks specifically for Taipei city plan. Moreover, Taipei city government also formulated “Taipei city multi-functional facilities open space design and maintenance regulation” in 1994. It aims to protect the quality and quantity of public spaces in the city and has been continuously revised to ensure that the designated spaces can satisfy the actual demand. Also, the city government implemented “Regulations of Bulk Reward policy” to encourage new developments to reserve more spaces for public uses. New urban planning districts’ planning policy has paid special attention to the percentage of the overall public spaces in the area and set requirements for sufficient setbacks from the street edges to be used as street-side public spaces throughout the area.

As urban morphology progresses from 1983, currently, Xinyi District (3) has become the most important commercial district in Taipei city, and the urban developments are expending continuously toward the eastern region (Nangang District (4)). Other commercial districts are mainly found outside of the central business area. To north-east is the Nanjing East road commercial district (5); to the south-east is Gongguan Commercial District (6); to the south-west is Wanhua district. In Wanhua district, urban functionality is less evenly developed where high-density developments are only at Ximending (7), and the south of Wanhua (8) has completely stopped growing. From the Rn analysis reading, the high Rn (Rn top 20%) is again corresponding with the major roadways, in particular to east–west arterial roads. These roads are also connecting all the major commercial districts in the city from Ximending (7), Taipei Main Station (9), Zhongxiao E. Rd. Commercial District (10), and Xinyi Commercial District (3). Theaters’ distribution is found to be in line with these major commercial districts location. The only expecting is at Ximending (7) where theaters are more densely populated because of the promotion through “Taipei Cultural and Creative Policy” (Ximending Film and Youth Creative Culture District). As for the low Rn (Rn bottom 20%) district, it is found to be mainly at the periphery of the city, and is also the area with less convenience accessibility level by public transportation and has low urban development intensity. In summary, the Rn analysis findings are accurately reflecting the current urban condition.

Analysis of urban functionality

In this research, we define the assessment categories by cross-referencing Taipei city’s Land-Use Regulations, literature reviews of related studies (Huang, 1995; Sawada et al., 1975) and the actual land use functions for organizing the criteria into 4 essential categories and 13 subcategories. The detailed list is shown in Table 1.

Table 1.

Urban function weighted scale assessment criteria.

Essential category/Definition	Sub-category	Spatial programs
R: ResidentialMainly residential use land	R1: Residential Use	Single-story mansion, villas, etc. luxurious residential areas
R: ResidentialMainly residential use land	R2: General Residential Use	General residential areas
C: Commerce	C1: Daily Commercial Use	Convenience stores, grocery stores, supermarkets, coffee shops, etc.
	C2: General Commercial Use	Clothing shops, home appliance stores, book stores, commercial offices, etc.
	C3: Specialized Commercial Use	Shopping malls, department stores, cinemas, large commercial offices, etc.
P: Public Facility	P1: Authorities Use	City hall, household affairs office, post office, etc.
	P2: Cultural and Educational Use	Primary schools, junior high schools, senior high schools, universities, etc., are open to the public on weekdays, nights and holidays.
	P3: Transportation Use	MRT station, transfer station, train station, etc.
	P4: Parking Use	Plane parking lot, solid parking lot, etc.
	P5: Market Use	Traditional market, market, etc.
O: Open Space	O1: Park Green Space	Parks, squares, scenic spots, etc.
	O2: Religious Facilities	Temples, churches, etc.
	O3: Dedicated Event Space	Stadiums, exhibitions, etc.

Weighted urban function scale

In this research, nine specialized experts were invited from the industry (three people), government sectors (three people) and Academia (three people). The experts from the industry are the directors or senior managers who hold a valid professional accreditation in their specialized fields. Experts from the government sector are directors from Taiwan’s National Spatial Plan division, Industrial Development Bureau and Regional Urban Planning division. From Academia, the experts are professors expertized in urban planning or related fields. All experts were invited to complete the AHP questionnaires to give scores on each urban function. Subsequently, these experts’ scores are processed through AHP analysis and to be used to compute the UFI for urban spaces (Saaty, 1980).

Category evaluated weighting (W1) matrix (Table 2) has a consistency index (CI) of 0.0225 (≤0.1) within acceptable deviation and a consistency ratio of 0.0250（≤0.1), which represents that W1 matrix has good consistency.

Table 2.

Urban function index pair evaluation matrix and weighting in essential category.

	R: Residential	C: Commerce	P: Public Facility	O: Open Space	W1
R: Residential	1	0.1765	0.2120	0.3521	0.0654
C: Commerce	5.6657	1	2.6085	3.7621	0.5270
P: Public Facility	4.7177	0.3834	1	1.3077	0.2393
O: Open Space	2.8403	0.2658	0.7647	1	0.1682

CI = 0.0225 (≤0.1); CR = 0.0250 (≤0.1).

CI: consistency index; CR: consistency ratio.

Category evaluated weighting (W2) matrix (Table 3) has a consistency index (CI) of 0.0437 (≤0.1) within acceptable deviation and a consistency ratio of 0.0280 (≤0.1) which represents that W2 matrix has good consistency.

Table 3.

Urban function index pair evaluation matrix and weighting in sub-category.

	R1	R2	C1	C2	C3	P1	P2	P3	P4	P5	O1	O2	O3	W2
R1	1	0.6287	0.2692	0.1598	0.1470	0.3420	0.2182	0.1470	0.2932	0.2582	0.2932	0.6934	0.2890	0.0183
R2	1.5907	1	0.5774	0.1740	0.1600	0.4174	0.2242	0.1554	0.3834	0.2371	0.2932	0.8807	0.3018	0.0224
C1	3.7141	1.7321	1	0.3521	0.3521	2.3956	1.0000	0.7230	2.5015	2.8403	1.5929	5.5034	1.3831	0.0868
C2	6.2573	5.7466	2.8403	1	0.5302	3.4597	1.9693	1.2523	2.7200	4.8568	3.0041	6.2488	1.7321	0.1536
C3	6.8041	6.2488	2.8403	1.8860	1	3.4597	2.0829	1.3636	2.6085	4.7177	2.5015	6.2488	2.2649	0.1736
P1	2.9240	2.3956	0.4174	0.2890	0.2890	1	0.7974	0.3192	2.0536	1.7100	0.8327	3.1326	0.9315	0.0572
P2	4.5826	4.4604	1.0000	0.5078	0.4801	1.2540	1	0.2932	1.7568	2.2649	1.9416	3.5569	1.1354	0.0820
P3	6.8041	6.4330	1.3831	0.7985	0.7334	3.1326	3.4110	1	4.7113	4.7177	3.0041	6.2488	2.9618	0.1641
P4	3.4110	2.6085	0.3998	0.3676	0.3834	0.4870	0.5692	0.2123	1	0.6555	0.6836	1.7831	0.9184	0.0448
P5	3.8730	4.2172	0.3521	0.2059	0.2120	0.5848	0.4415	0.2120	1.5254	1	0.7647	4.5180	0.6287	0.0474
O1	3.4110	3.4110	0.6278	0.3329	0.3998	1.2009	0.5150	0.3329	1.4629	1.3077	1	4.7786	1.4239	0.0647
O2	1.4422	1.1354	0.1817	0.1600	0.1600	0.3192	0.2811	0.1600	0.5608	0.2213	0.2093	1	0.3376	0.0207
O3	3.4597	3.3133	0.7230	0.5774	0.4415	1.0735	0.8807	0.3376	1.0889	1.5907	0.7023	2.9618	1	0.0644

CI = 0.0437 (≤0.1); CR = 0.0280 (≤0.1).

CI: consistency index; CR: consistency ratio.

Table 4 lists the sequence order for weighted index categories. For W1, the order is Commerce (52.7%) > Public Facility (23.9%) > Open Space (16.8%) > Residential (6.5%); and W2’s essential categories values (which is the sum of the sub-category’s values) order is Commerce (41.4%) > Public Facility (39.6%) > Open Space (15%) > Residential (4.1%). W1 and W2 have the same categorical orders and similar weighted values with coefficient values that is W1 (79.2%) < W2 (73.7%). We believe the reason for this result is because W2 values are based on the comparison between detailed sub-categories which can result in more comprehensive findings. The consistency ration of the hierarchy (CRH) of evaluation categories and indexes is 0.0276 (≤0.1) which falls within the acceptable range. Therefore, the analysis of UFI for the historical times can be analyzed based on W1 categories, and for 2018’s complex urban conditions, the analysis is based on W2 categories and sub-categories for more detailed and comprehensive analysis and results.

Table 4.

Urban functionality categories and index weighting sequence.

Importance sequence	Category weighting value（W1）			Category weighting value（W2）
Importance sequence	Category	Weight	%	Category	Weight	%
1	C: Commerce	0.5270	52.7	C3: Specialized Commercial Use	0.1736	17.4
2	P: Public Facility	0.2393	23.9	P3: Transportation Use	0.1641	16.4
3	O: Open Space	0.1682	16.8	C2: General Commercial Use	0.1536	15.4
4	R: Residential	0.0654	6.5	C1: Daily Commercial Use	0.0868	8.7
5	–	–	–	P2: Cultural and Educational Use	0.0820	8.2
6	–	–	–	O1: Park Green Space	0.0647	6.5
7	–	–	–	O3: Dedicated Event Space	0.0644	6.4
8	–	–	–	P1: Authorities Use	0.0572	5.7
9	–	–	–	P5: Market Use	0.0474	4.7
10	–	–	–	P4: Parking Use	0.0448	4.5
11	–	–	–	R2: General Residential Use	0.0224	2.2
12	–	–	–	O2: Religious Facilities	0.0207	2.1
13	–	–	–	R1: Residential Use	0.0183	1.8

CIH = 0.8440; RIH = 30.6179; CRH = 0.0276.

CIH: consistency index of the hierarchy; CRH: consistency ration of the hierarchy; RIH: random index of the hierarchy.

From Table 4, C3: Specialized commercial use, P3: Traffic transfer use and C2: General commercial use are the highest scored categories; these three categories fall in the first zone functionality with an average of 16.4% weighted scale. These urban functions are often generating large assembly of crowds (e.g. department stores, movie theaters, shop malls, of C3; MRT stations, Transit stations, Railway stations of P3; Clothing stores, Appliance stores, Bookstores, Business Offices of C2); therefore, it has the highest capability to support commercial or social activities to occur. R2: General residential use, O2: Religious facilities, R1: Pure residential use are the three categories that scored the lowest; they fall in the third zone functionality with an average of 2% weighted scale. Especially R1 and R2 are dedicated to residential uses where only casual social events might take place, but with very minimal commercial activities allowed, therefore, social activities are less likely to occur there. O2: Religious facilities has been an essential facility in the past as a community center (Tseng, 2011) which hosts traditional Taiwanese Opera, Glove Puppets or large festive events. It is also a place where food stalls and restaurants would gather around (e.g. night markets, temple fairs) for the substantial pedestrian traffic. However, over time, these urban functionalities are gradually replaced by modern facilities; outdoor performances are replaced by modern theaters, street vendors are replaced by restaurants and supermarkets. Today, O2 is a place that hosts religious events and a meeting point for a few local elderlies.

Urban function index analysis

UFI is an indicator of the urban functional strength which is driven from the quantitative calculation of expert’s opinions with the AHP method. In this research, we divide Taipei city into 123 smaller blocks for calculation based on two criteria. Firstly, the block is defined by the Choice analysis findings where the high value indicates the major arterial roads with high traffic flow and wider streets that naturally became a barrier for pedestrian flow (Koohsari et al., 2013). Secondly, the block dimension is controlled between a radius of 250 and 1000 m. The effective distance for amenities of a life circle is at the max of 1000 m which is approx. 17 min by walking (Prinz, 1997). The greater the radius is, the less willingness for people to approach. However, when the block is smaller than 250 m, it becomes difficult to accommodate amenities like green spaces and daily commodity services to form a complete living circle. Lastly, it is also important to limit the difference in each block size to minimize discrepancy in the research outcome.

This research compares 2018 Taipei city’s Rn values and UFI findings, and based on their ranking divides the districts into four quadrants and maps the summarized result as shown in Figure 2. The quality of each quadrant is as follows:

The first quadrant: High UFI and high Rn districts are Ximending C.D. (1), Dadaocheng C.D. (2), Taipei Main Station C.D. (3), Eastern C.D. (4), Xinyi C.D. (5), Songshan Station C.D. (6), etc. These are the districts that have high-density urban function and are conveniently accessible. These districts are currently the most significant commercial districts of Taipei city and located along arterial roadways. Additionally, there are always MRT stations located in proximity to support transportation convenience.

The second quadrant: High UFI and low Rn districts are Longshan Temple (7), Yuanshan (8), Gongguan C.D. (9), etc. These are the districts that also have high-density urban function but are not conveniently accessible. Gongguan C.D. is one of the major commercial districts and Yuanshan and Longshan Temple are considered as extensions of major commercial districts (Yuanshan is extended from Zhongshan North Road C.D., Longshan Temple is extended from Ximending). Although, these districts are located relatively to the periphery of the city where it is not as conveniently accessible. However, all these three districts have MRT stations located closely to support the transportation convenience.

The third quadrant: Low UFI and high Rn districts are Mingsheng District (14), Taipei Winery (15), Taipei Machine Factory (16), Yanji Street (17), and Linyi Street (18), etc. These areas have low-density urban functions but are conveniently accessible; in other words, these districts are inefficient with their land utilization rate. These districts are mostly populated with deteriorated old buildings that require urban renewal plans or are the most discussed areas for urban function transformation topics.

The fourth quadrant: Low UFI and low Rn districts are South of Wanhua district (10), Wuxing Street (11), Nangang Park (12), Linguang (13), etc. These are the areas that have low urban function intensity and are not conveniently accessible. These areas are mainly residential blocks located at the outskirts of the city as part of the suburbs of Taipei city. Only Nangang Park and Linguang have MRT stations to support the limited suburban street networks. South of Wanhua and Wuxing Street had no MRT system support, and service mainly by buses which can bring down the accessibility level of the area. However, in the near future, there will be MRT Wanda Line to service South of Wanhua (10) and will set up LG02, LG03, LG04 stations within the district. Also, Taipei city government has started planning new south-east MRT for Wuxing Street (11) area and Y39 station too.

Figure 2.

Spatial configuration and urban functionality comparison overlay (2018).

Based on the above analysis results, there are plans to set up MRT stations regardless of the districts’ high and low UFI or Rn characteristics. This finding is in line with the mass transportation oriented urban development model (TOD) that Taipei city was based on, where the stations are the center of regional development for planning high-intensity and mix-use urban functions. In this research, we apply space syntax methodology to analyze urban spatial configuration and make assumptions on accessibility levels. With higher convenience level, the development intensity shall be greater too. However, spatial configuration is not the sole factor of urban development intensity. Based on quadrant analysis, we learn that not all districts are consistent with the space syntax analysis assumptions. For example, the second quadrant (high UFI, low Rn) districts’ actual urban planning can significantly improve convenience level through the MRT stations. From our research analysis of second quadrant districts, MRT stations are identified in proximity for all these areas; therefore, the actual convenience level is high. Additionally, there are other factors that make urban development limited and unable to update for third quadrant (low UFI, high Rn) districts; from the interview with the urban planning experts (Taipei City Urban Design Review Committee) we gain insight that some areas are populated with deteriorated residential buildings over 30 years old but the lack of incentives for urban renewal has discouraged the private sectors to take action, e.g. Minsheng District, Yanji Street and Linyi Street, etc. Another reason has been that part of this area has historical industrial buildings and is preserved for its historical significance where the zoning has remained as industrial with low urban function intensity; therefore, the UFI analysis results in lower values, e.g. Taipei Winery, Taipei Machine Factory, etc.

In this research, we refer to the overlay analysis method of urban functionality (UFI) and convenience level (Rn values) for 2018 as the model to analyze the urban functionalities of Taipei city’s changing developments over time. However, due to the limited historical data and simpler land-use categories which differ from today’s complex urban functions, we apply Table 2’s weighted values (W1) with four essential categories (C: Commerce, P: Public Facility, O: Open Space, R: Residential) and refer to relevant documentation to make land-use pattern assumptions and perform overlay analysis (Figure 3).

Figure 3.

Urban spatial configuration and urban functionality comparative analysis (1895, 1907, 1945, 1983, 2018).

From the observation of 1895’s overlay analysis, we can clearly identify the boundary of Taipei city’s “three market-streets” (Monga, Dadaocheng, and Taipei city): the higher UFI blocks are the main streets at the time and it is also in line with the high Rn street locations. The only exception is the area where the junction of outside Taipei city and Dadaocheng meets. This area has low UFI and high Rn values. We believe the reasons as follows: it is a major connector road between Taipei city and Dadaocheng, and Taipei Main station is also housed here where the railway crossing leads to the division of the space. Both of the reasons contribute to the difficulty for the crowds to converge in this area, therefore except the station area, the rest of the areas were a wasteland in 1895.

For 1907 overlay analysis, the “three market-street” boundary has become indistinctive, but it is still identifiable from the UFI mapping that Taipei city, Dadaocheng, and Monga were the central development areas. The station area remains the same as 1895’s reading; the area stays divided by the railway crossing. There were only a few residential buildings in this district; one particular cluster at the south of these areas (Guting) has a high UFI (60%–80%) value. This is a newly developed district in 1907, which includes school, markets and mix-use developments, therefore, reflecting with the higher UFI result and is consistent with the changes of urban developments in history.

For 1945 overlay analysis, the urban developments continued from 1907’s spatial configuration, high UFI and high Rn districts were still at Dadaocheng, Monga, Taipei city and Guting, etc. These areas remained as the central districts of Taipei city. However, the developments have begun to shift to North and East regions slowly, but due to the short development period, the new districts only have the sporadic distribution of buildings and low-density urban functions resulting in low UFI finding.

For 1983 overlay analysis, the urban configuration is a continuation from 1945, but the city was expanding rapidly. The reasons are as previously stated in the Analysis of spatial configuration section. After 1949, the vast influx of immigrants from mainland China has increased the demand on land uses drastically. High UFI and high Rn districts are at Dadaocheng, Manga, Taipei city, Guting and Zhongxiao East Road Commercial District; these areas were the central areas of Taipei city. Although Xinyi Plan District has already been in place, however, due to the short development time, there were only Governmental agency facilities in the area which results in low UFI finding.

Together, space syntax methodology and the UFI analysis model can shed insights on the changes in urban development contexts of different times in history. The sequence of urban construction is to introduce roadway systems first and plan urban functions accordingly; later, the Government and the private sectors can begin to develop. In the context of urban morphology of different times in history, we learn that when the street networks are in place for a newly planned district, the high Rn districts in the city will begin to shift toward the new area. Next, we can observe the new districts begin to develop with higher UFI result. Therefore, the changes in Rn values can be a predictive tool for early urban developments. Over time, the high Rn districts will begin to have higher UFI readings. However, from our overlay analysis of Rn and UFI for Taipei city in 2018, we learn that not all the districts are consistent with our assumption of high Rn and high UFI districts locations because integrated level (Rn) is not the sole factor of the urban developments. UFI analysis parameters are mainly based on the overall floor area, urban function weighting and other factors of limitation like regulation and lack of reward incentives have discouraged residents to take any actions to develop (e.g. Minsheng District). Thus, these can result in limited overall developed floor area usage which leads to lower UFI result. Other factors can be based on the original low land uses intensity as industrial or military (e.g. Taipei Winery) and have not been transformed for new proposed uses yet. Thus, could also result in low urban functionality weighting and lower UFI findings. Moreover, the TOD oriented urban development model can affect Rn and UFI overlay result too, for example, mass-transit stations are beneficial to the convenience level of a district. Therefore, the lower Rn district (from space syntax analysis) can have high actual convenience level through the support of approximate mass-transit stations. Based on our findings, the low Rn and high UFI districts all have nearby MRT stations in place thus proving that mass-transit station can make districts with lower convenience level (Rn) areas still have high UFI opportunities.

Discussion

This research applies space syntax methodology to analyze Taipei city’s urban spatial configuration and urban morphology from 1895. Simultaneously, it observes the distribution of the “common urban venues” (theater, MRT stations, Starbucks) for social congregations and commercial activity patterns to investigate the transformation of urban developments over time. Additionally, this research also utilizes AHP on expert’s opinion to attain scores of different urban functions and perform the weighted calculation to attain the UFI for all Taipei city blocks. Lastly, we overlay the results of both Rn and UFI to obtain insight into their correlation. The research findings are summarized as follows:

Based on the syntactic values between 1895 and 2018, both the Global Integration findings have coincided with the actual city developments over time. In 1895, Rn values were highest at the ‘three market street’ districts. Particularly, the highest Rn value is at the land-water transport interchange where it was the most active area at the time. In 1907, after the Japanese’s urban rectification plan, city street network had transformed from organic to grid pattern, and Taipei city began to develop more evenly. The findings are very different from 1895’s analysis where the highest Rn is much concentrated. The high Rn value has shifted to the relatively new arterial roads indicating that Taipei city has grown into five major commercial districts (Taipei city, Schools, Monga, Guting, Dadaocheng). In 1945, the urban spatial configuration had remained unchanged since 1907 but extended further outward. Due to geographical limitations, the city has mostly expanded toward the east, and the Rn analysis can reflect the city’s migration. Rn has a similar reading from 1907 which implies that street network structure has minimal changes, but only the higher reading has shifted eastward. Today, although Taipei city’s spatial network structure remained similar to 1945’s spatial structure, the city has been expanding toward east very quickly. It is due to the massive influx of immigration moving into the city since 1949 and has currently reached a saturated status. Higher Rn findings have stayed in similar locations to the previous year’s finding, and lower Rn readings are at the city periphery (South of Wanhua, Wuxing St, Nangang, etc.). Comprehensively, Rn analysis findings have met the city’s historical development over time which demonstrated reasonable ability to comprehend the urban development (Tsou and Huang, 2009).

Furthermore, we observed a relatively low correlation (eco-efficiency 0.147) between the Rn and UFI analysis findings. We believe the main reasons are due to the following factors: from our comparative study of Taipei city’s urban development and current urban condition, it is clear that Rn reflects the development and changes of the urban area whereas UFI relates to the intensity of an area’s urban functions. However, there are other circumstances that affect the development process of the city where official urban planning and policies are often unable to reflect these changes in time. In particular, the urban renewal process is one of the major determinants, for instance, when urban planning is outdated with un-revised urban functions skewing the analysis finding as each urban function is weighted differently. In addition, outdated urban planning also does not indicate accurate buildable total floor area which can lead to the inaccuracy of our results. The above two conditions are often seen at Industrial facilities or old residential building neighborhoods where due to the misrepresented inputs we are observing clear inconsistency reading between UFI and Rn. It is based on these reasons discussed that we further applied the quadrant analysis to our result to pinpoint the distinct “inconsistent” areas (e.g. UFI and Rn findings are uncorrelated) and further investigate the underlying problems to better inform future urban renewal decisions.

We overlay urban functionality and space syntax findings and summarize the urban district characteristics into four quadrant groups. Also, the study includes interviews with Taipei city urban planning experts (Taipei City Urban Design Review Committee) for their knowledge regarding urban history and developments. The analysis and interview results show that the four-quadrant findings are consistent with the actual urban conditions. The first quadrant is the high UFI and high Rn districts which represent these are the areas with high-density urban functions and high convenience level. These districts are currently the most significant commercial districts of Taipei city, including Ximending C.D., Taipei Main Station C.D., and Xinyi C.D., etc. The second quadrant is high UFI and low Rn districts which represent that these districts also have high-density urban functions but low convenience level. In actual condition, these areas are still the major commercial districts or a part of the major commercial districts that have intense urban functions. But because of these districts are at the edge of the spatial network in this study, therefore, the calculate results are in low integration values. However, Taipei city was planned as a TOD oriented city where the MRT stations are to be set up along with the major roadways and supported by other modes of transportation (friendly pedestrian sidewalks, public bicycles, public electric vehicles, buses) to increase the accessibility level. All these districts have MRT stations in proximity to remediate the low convenience level due to the limitation of the distant locations. The third quadrant is low UFI and high Rn districts which represents that these districts have low urban functionality but high convenience level. These districts are mainly populated with deteriorated old buildings that need urban renewal desperately, or are the most talked about areas for land uses transformation plans (e.g. Taipei Winery, Taipei Machine Factory, Yanji Street, etc.) UFI analysis parameters are based on urban functionality weights and total floor area because land uses (affect the urban functionality weighting) and FAR (affect total floor area) can both affect the result of the UFI calculation. These third quadrant districts mostly remain with their old low-intensity land use and are not updated by land use transformation plan which can be the reason to result in lower UFI values. In the future, urban renewal can potentially change these areas’ land uses, or increase the FAR, which can contribute to having a positive effect on the UFI analysis. The fourth quadrant is low UFI and low Rn districts which represents that these districts have low-intensity urban functions and low convenience level. The areas are mainly residential blocks located at the outskirts of the city as part of the suburbs of Taipei city, such as Wuxing Street, Wanhua South, etc. Again, the above finding is in line with the actual urban conditions. The above analysis results prove that this methodology can reasonably interpret the advantages and disadvantages of urban functionality and convenience level. Therefore, this methodology can be used to quickly compile the pros and cons of each district. Subsequently, the authorities can focus on the identified urban development issues for a more detailed and efficient discussion in making or revising the urban policies.

Through applying both space syntax and UFI evaluation to analyze Taipei city’s urban development, the results have indicated proximity to reflect the actual urban condition for the investigated period. Thus, it validates our proposed analytical framework as a useful tool in helping the future urban planning processes. This comparative analysis can efficiently reveal the condition of the different districts and bring attention to the strengths and weaknesses of the area. Building from these preliminary findings, the stakeholders can efficiently pinpoint the challenges and issues of the current condition and make informed decisions on the future developments or improvements plans for the city.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is a part of the research outcome of the project “A study of the influence of urban planning on the change of social networks by means of the method of space syntax and social network analysis” (MOST 106-2410-H-845-016-MY3), which was funded by the Ministry of Science and Technology, Taiwan. We thank the Ministry for funding this project.

Yin-Hao Chiu, PhD, is from School of Architecture and Built Environment, University of Nottingham. Currently, Professor Chiu is the Dean of the School of City Management, University of Taipei. He also serves as the general consultant of both local and central public sectors in the fields of urban planning, urban design, and environment as well as landscape. His research interests include eco-city and green building, physical environment, urban studies, urban development and policy, wind engineering and environmental behavior.

I-Ting Chuang is currently a Doctoral Student of architecture and sustainable design, Singapore University of Technology and Science, and was previously an Assistant Professor at Architecture Department, Taiwan Tech, Taiwan. Ms Chuang was educated at University of Auckland, New Zealand, and graduated with first honor B. Arch. Then she attended Graduate School of Design, Harvard University, USA, gaining her MDesS. Ms Chuang spent 7 years in New York practicing Architecture in renown firms before embarking on an academic career which focuses on architecture design and studio curriculum. Ms Chuang’s current research focus is on design informatics in urban geography with emphasis on the spatial quality of urban spaces.

Chi-Yao Tsai has Master from the Graduate School of Architecture and Urban Design, Chinese Culture University, Taiwan. Tsai also serves as a Research Assistant of Department of Urban Development, University of Taipei. Tsai’s research interests include urban studies, wind engineering, and physical environment.

ORCID iD

I-Ting Chuang

References

Hillier

(1996) Space is the Machine: A Configurational Theory of Architecture. Cambridge, New York: Cambridge University Press.

Hillier

Hanson

(1984) The Social Logic of Space. Cambridge: Cambridge University Press.

Hoeven

van der and Akkelies van N (2014) Improving the design of urban underground space in metro stations using the space syntax methodology. Tunnelling and Underground Space Technology 40: 64–74.

Huang

(1995) A study for the commercial spatial structure in Taipei. Land Forum 17: 51–66.

Koohsari

Kaczynski

Giles-Corti

, et al. (2013) Effects of access to public open spaces on walking: Is proximity enough? Landscape and Urban Planning 117: 92–99.

Lee

Lin

(2007) A space syntax analysis of settlements in Taiwan. Journal of Architecture 60: 27–45.

Liang

(1999) A study on spatial structure and functional activities of business areas in Taichung. Bulletin of the Geographical Society of China 27: 153–182.

Lin

(2015) Cities ecology trace-Revisited the historical characteristics of Taiwan’s traditional urban space. Journal of Culture Resources 9: 47–83.

Peponis

Ross

Rashid

(1997) The structure of urban space, movement, and co-presence: The case of Atlanta. Geoforum 28: 341–358.

10.

Prinz W (1997) Perception and Action Planning. European Journal of Cognitive Psychology 9: 129–154.

11.

Saaty

(1980) The Analytic Hierarchy Process. New York: McGraw-Hill, Inc.

12.

Sawada

Tanaka

Shirai

(1975) The geographical study of CBD-ratio and areal structure on Sendai city. Research Bulletin of the Institute of Natural Science of Nihon University 10: 31–42.

13.

Shih

(2017) The cultural heritage maps of the old theaters in Taiwan (1895-1945). Available at: http://map.net.tw/theater// (accessed 29 April 2018).

14.

Tseng

(2011) From private dining halls to drinking parlors: The formation and expansion of public spaces in modern Taiwan (1895-1911). Journal of Chinese Dietary Culture 7(1): 89–142.

15.

Tsou KW and Huang SW (2009) Space Syntax Theory Applied to the Relationship between Street Configuration and Spatial Distribution of Urban Commercial Development. City and Planning 36(1): 81–99.

16.

Tzeng

Chen

Lin

(1995) The establishment standard for park and greenery area in the urban planning of Taipei city. Journal of City and Planning 22(2): 93–107.

17.

Wang

Mao

Dang

(2008) An evolution model of Beijing – A discussion on the evolution of urban spatial and functional patterns based on space syntax. Urban Planning Forum 3: 82–88.

18.

Yin

Chen

(2016) The imaginary of architectural modernity and its termination: The birth and death of Zhunghua shopping yard. Journal of Architecture 96(S): 1–26.